Google Gemini API Gemini Pro Ultra Flash Integration

Google Gemini API Integration: Gemini Pro, Ultra, Flash

Google Gemini is natively multimodal: processes text, images, audio, video and code in a single context. Gemini 1.5 Pro has a context window of 1 million tokens — a unique characteristic for working with large documents. Gemini Flash is fast and cheap for high-load tasks.

Basic Integration via Google AI SDK

import google.generativeai as genai
from google.generativeai.types import HarmCategory, HarmBlockThreshold

genai.configure(api_key="GOOGLE_API_KEY")

model = genai.GenerativeModel("gemini-1.5-pro")

# Simple call
response = model.generate_content("Explain quantum computing")
print(response.text)

# Generation configuration
response = model.generate_content(
    "Data analysis",
    generation_config=genai.GenerationConfig(
        temperature=0.1,
        max_output_tokens=2048,
        response_mime_type="application/json",  # Force JSON
    ),
)

# Multimodality: text + image
import PIL.Image
image = PIL.Image.open("diagram.png")
response = model.generate_content(["Describe the architecture in the diagram:", image])

# Video analysis (unique to Gemini)
video_file = genai.upload_file("presentation.mp4")
response = model.generate_content(["Make a summary of the video:", video_file])

Streaming and Async

# Streaming
for chunk in model.generate_content("Long text...", stream=True):
    print(chunk.text, end="", flush=True)

# Async
import asyncio

async def async_generate(prompt: str) -> str:
    async_model = genai.GenerativeModel("gemini-1.5-flash")
    response = await async_model.generate_content_async(prompt)
    return response.text

Chat with History

chat = model.start_chat(history=[])

response = chat.send_message("Hello! My name is John.")
response = chat.send_message("What is my name?")
# Model remembers context from history

Function Calling (Tool Use)

def get_stock_price(ticker: str) -> dict:
    """Returns stock price"""
    return {"ticker": ticker, "price": 150.0, "currency": "USD"}

tools = [get_stock_price]  # Gemini accepts Python functions directly!

model_with_tools = genai.GenerativeModel("gemini-1.5-pro", tools=tools)
response = model_with_tools.generate_content("What is the price of Apple (AAPL)?")

Vertex AI (Enterprise)

import vertexai
from vertexai.generative_models import GenerativeModel

vertexai.init(project="my-project", location="us-central1")
model = GenerativeModel("gemini-1.5-pro-preview-0514")
response = model.generate_content("Request")

Cost of Gemini (2025)

Timeline

• Basic integration: 0.5–1 day
• Multimodal scenarios: 2–3 days
• Vertex AI production: 1 week